matlab simulation of solar panel msx-64 at the best...

energyequipsys/ Vol 4/No1/June 2016/ 21-30

Energy Equipment and Systems

http://energyequipsys.ut.ac.ir www.energyeuquipsys.com

Matlab simulation of solar panel MSX-64 at the best locations of Kermanshah province using GIS interpolation

Authors

Milad Imani-Harsini a*

Mohammad M. Karkhanehchi

b

a Department of Electronics, College of

Engineering, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran; Department of Electronics, College of Engineering, Kermanshah Science and Research Branch, Islamic Azad University, Kermanshah, Iran. b

Department of Electronics, Faculty of

Engineering, Razi University, Kermanshah, Iran; Department of Electronics, College of Engineering, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran

ABSTRACT

Considering that the effective yield of a panel is equal to its total number of hours of solar radiation and temperature, only the effects of temperature and solar radiation intensity at the maximum power point (MPP) are investigated in this article. By collecting temperature data, sun's radiation hours from six synoptic meteorological stations in Kermanshah Province over the course of an eleven-year period (1995-2005), with the use of GIS software, a map of Kermanshah Province's temperature and radiation based on plotted latitude and longitude as well as the establishment of regression, the most suitable location for solar panels is proposed. In this MATLAB software simulation using the characteristics of panel MSX-64, all parameters have been considered and determined by the characteristics of the panel. Throughout the process, the design has been based on four parameters as the primary specifications of the solar panels: Isc, Voc, Imp, Vmp. With the ability to simulate other solar panels with temperatures and radiation intensity corresponding to each area, the I-V curve of each custom solar panel can be drawn, making it possible to obtain the maximum power.

Article history:

Received :7 November 2015 Accepted : 9 January 2016

Keywords: GIS, Interpolation, MATLAB, Regression, Solar panel Msx-64.

1. Introduction

As a result of population growth and increasing industrialization, the demand for renewable energy is increasing every day [1]. In turn, the importance of “green”, environmentally-friendly, non-polluting, renewable energy sources have greatly increased. The sun is, and will remain, our main source of lasting energy for the next 5 billion years [2]. The mass of the sun is 333 *Corresponding author: Milad Imani-Harsini Address: Department of Electronics, College of Engineering, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran; Department of Electronics, College of Engineering, Kermanshah Science and Research Branch, Islamic Azad University, Kermanshah, Iran E-mail address: [email protected]

thousand times larger than that of the Earth, and is converted into energy at a rate of 4.2 million tons per second [3]. In order to harness the power of solar energy, we must use solar cells which are electronic devices and use photovoltaic phenomenon for conversion of light or photon into electrical current or voltage [4].

If just 0.1% of the Earth’s surface was covered with solar panels and had only 10% efficiency, this would be enough energy fulfill all human requirements [5]. Since the efficiency of each solar cell is usually low, combine both series and parallel arrays so as to enrich the ultimate performance and expected power. Due to their efficiency, receiving the highest possible level of

http://energyequipsys.ut.ac.ir/

22 Milad Imani-Harsini & Mohammad M. Karkhanehchi /energyequipsys / Vol 4/No1/June 2016

efficiency in a solar panel is the most important topic in each type. The maximum power point (MPP) [6], and the corresponding current and voltage (Vmp, Imp) can determine the performance of a panel and the basic factor for naming it [7]. For example, MSX-64 is a panel with maximum power of 64 watts. Since the non-linear performance characteristics of the I-V and P-V curves is a function of temperature and radiation [8], it is essential to incorporate these parameters in analysis. 2.Materials and Methods The List of symbols and terminology used in this article can be observed in Table 1.

2.1. Temperature and radiation maps By analysing data from a six synoptic station Kermanshah, Kangavar, West Islamabad, Sarpolzohab, Ravansar, and Sararood in the eleven-year period mentioned, the results

obtained about temperature range are shown in Table 2.

With the purpose of accessing the radiation map by checking the hours of sunshine data in the range of the six stations mentioned, the results in Table 2 were obtained.

By introducing temperature data obtained from Table 2 and the latitude and longitude, respectively the five level analysis and interpolation by using GIS software in Fig.1a is observed. By introducing sunshine hours’ data obtained from Table 2 and the latitude and longitude, respectively the five level analysis and interpolation by using GIS software in Fig.1b is observed.

As Fig.1a shows, the eastern regions are the hottest regions and the western regions with the restricted area of the southern part of Kermanshah are the coldest regions in this province. Then, by observing Fig.1b, it is concluded that, in Kermanshah province, in the eastern regions and some limited areas in the West, the highest part of radiation in the location can be found.

Table 1. List of symbols and terminology

Symbols Description Symbols Description

Ipv The radiation flow Rs Series resistance

Voc Open circuit voltage Rsh Shunt resistance

Isc Short circuit current s Irradiance

Vmp Maximum power of voltage k Boltzmann constant 1.38 × 10-23 (joules per kelvin)

Imp Maximum power of current q Electron charge 1.6 × 10-19 (colon)

T Temperature α Ideal factor

Tref Reference temperature n Emission factor

Table 2. Average temperature and number of sunshine hours per month by each station in the eleven-year period from 1995 to 2005

Name of station Temperature (in degrees Celsius) Average solar radiation (in hours)

Kermanshah 15.48 254.10

West Islamabad 13.87 263.48

Kangavar 13.12 260.15

Ravansar 15.18 259.71

Sarpolzohab 20.20 263.64

Sararood 14.70 253.83

Milad Imani-Harsini & Mohammad M. Karkhanehchi /energyequipsys / Vol 4/No1/June 2016 23

Fig.1. Kermanshah province maps: a) Temperature map in Kermanshah province in the eleven year period from 2005 to 1995; b) Radiation map in Kermanshah province in the eleven year period from 2005 to 1995


2.2. Modeling

In general, because solar cell is based on P-N junction, being exposed to light radiation, it can model a cell with diodes, which uses source flow instead of radiation [12].

In Fig.2a, the equivalent circuit of a polycrystalline silicon solar cell [13] is observed.

The curve I-V and P-V characteristics and effects series and parallel resistors on the performance of the solar cells [14] in Figure 2b are observed.

As shown in Fig.2b, increasing the series resistance adversely affects maximum power; therefore, a resistance of the lowest possible value (close to zero) is optimal [15]. The opposite is true for parallel resistance, whereby its reduction lowers the maximum power point; therefore, the greatest possible value of parallel resistance (close to ∞) is ideal [16]. According to what was mentioned in the introduction, the Maximum Power

Point (MPP) dependence on temperature will be demonstrated in this simulation.

The MSX-64 solar panel is an upgrade of the basic MSX-60 model. Its functional improvements [17], as presented in Table 3, along with its associated costs in relation to other panels, are fundamental considerations for its selection.

The correlations between the maximum voltage point, Vmp, and intensity of the radiation, s, as well as ambient temperature, T, was established based on [18], [19]

Vmp = f(s,T). (1)

Because the objective is to achieve the maximum values of the voltage and current (Vmp and Imp) as based on Fig.2b and Table 3, relation

Ipv = I – I0 (

(2)

is derived.

Fig. 2. Solar cell: a) equivalent circuit of a solar cell; b) I-V and P-V characteristic curves of a solar cell

Table 3. Characteristics of solar panel MSX-64

Data value

Maximum power (Pmax) 64 W

Maximum voltage )Vpm) 17.5 V

Maximum current )Ipm) 3.66 A

Short circuit current )Isc) 4 A

Open circuit voltage )Voc) 21.3 V


Equation (2) [20] shows that the ideal output of a photovoltaic cell, denoted I0, is the reverse saturation current of the diode. In applying Eqs. (1) and (2), a simulated model of the temperature and radiation is created. Note that DI, C1, C2, and VR are defined quantities used to simplify the final equation. That is

C2 =

(3)

C1 = (1–

)(

)

(4)

DI = sα(T–Tref) + Isc(s–1) (5)

VR = Vpv + (T–Tref) + RsDI (6)

Coefficients α and β are defined according to the following equations:

= Isc × 0.0012 (7)

= Voc × 0.005 (8)

Applying Eqs. (3), (4), (5) and (6) in tandem generates the conclusive equation:

Ipv = Isc (1– [C1

) –1]) + DI

(9)

The simulation of the resulting system of equations is illustrated in the figure below.

Circuit blocks B, C, D, E, and F in Fig.3 of the simulated model are subsystems of the solar panel (Fig.3; block A). In this simulation, the input values are temperature and radiation intensity (Fig. 3; block A), and directly influence the IPV output value (Eq.(9); Fig. 3; block F).

As previously mentioned, a change in the values of the four basic parameters ISC, VOC, IMP, and VMP will alter simulation results for each other panel (Fig. 3; block B). The values in effect for the MSX-64 panel simulation are listed in Table 3.

The simulation was executed with temperature values of 25 and 50 degrees Celsius and a radiation intensity of s=1 (1000 w/m2). Once again the simulation was conducted, with a variation in radiation intensity, at values of s=0.5 and s=1 (500 and 1000 w/m2 respectively) and a temperature of 25°C.

As seen through Eq.(1), temperature directly influences the performance of the

Fig. 3. Solar panel simulation using MATLAB software


Fig. 4. Solar panel simulation: a) results at temperatures of 25°C and 50°C; b) results of simulated solar radiation at 500 and 1000 watts per square meter


solar panel. We expected that a temperature change would cause slight variations in IMP, and substantial changes in VMP. The end goal is to determine the peak performance level of the solar panel, which according to Fig.4a, is achieved at a lower temperature.

Equation (1) suggests that a change in radiation intensity has a major impact on the performance of a solar panel. Figure 4b shows that while these changes have no effect on VMP, they greatly affect IMP, which is the more influential variable on the maximum power point. Therefore, greater radiation intensity results in the improved performance of the solar panel.

2.3. Regression

While the temperature and radiation maps of Kermanshah and the results of the simulation as well as sunshine hours are major factors in the intensity of the radiation, with the establishment of the regression between temperature and the sunshine hours by using IDRISI software, the temperature can be calculated on the basis of the sunshine hours and the best location for the placement of solar panels can be determined in Kermanshah and run the simulation there.

Regression based on the regression Eq.(10) has been performed in Fig.5. In this regard, R represents the multiple correlation coefficients between the dependent variable and the independent variables. The value of apparent

R in the regression equation is equal to 0.73 and R2 which shows the variability of the dependent variable based on the independent variables, 54 percent of the station's temperature is demonstrated by the independent variable. The adjusted R and R2 represent the amount of corrective effects related to the number of variables; these coefficients show modified complexity of the model. After adjusting the effects of the regression equation according to the number of variables, a significant change in the values of R and R2 cannot be seen.

In this regression equation, the following critical values to check significant review of the equation have been used. The regression coefficient has the F-statistic equal to 127765.53 at a significant level of 99% with degrees of freedom 3.0173= 105748, which suggests this equation is significant at 99% level.

T-test and ANOVA table are also in Fig. 5 where they can be observed. T-statistic and F-statistic combine the most common type of test to estimate the relative success of the model and the possibility to add or delete independent variables in the regression model. Coefficients indicate the separated effects of each independent variable over the dependent variables. The meanings of coefficients are expressed in the T-test. T-test reveals that each significative independent variable moves from zero (no effect) in the separated regression table where 72244 degrees of

Fig. 5. Results of regression


freedom are listed. In this regression equation, the following critical values to check significative reviews of the equation were used. It has a significant level of 99% accuracy with 2.326= 105748 degree of freedom.

ANOVA table is used to show the acceptability of the model in statistical terms. Regression line in the ANOVA table shows the information about change in the model. Column mean square shows parts explained by the regression. These amounts are obtained from subtracting the sum of remaining squares out of the total sum of squares. Residual analysis could help us to feature the model which is identified in a specific area. Remnants of the positive value of the regression equation show that the model can predict values of temperature lower than the actual amount. That is

Temperature = 0.1167 + 0.5219 (Sun shine).

(10)

3.Results and Discussion Considering the regression results and the simulation outcomes, the most appropriate location for the placement of the solar panels are proposed.

Figure 6 indicates that prime locations for the solar panels are in Kangavar and West Islamabad. Based on Fig.7a and the results from Fig.7b, the environmental conditions of the prime locations have been investigated. The maximum power value of the MSX-64 panel is compared with the reference values in Table 3.

The average monthly temperature of the most favorable panel locations taken over the eleven-year period is 13.87 degrees Celsius, with minimum radiation intensity (yearly average during winter solstice) of 861 watts per square meter. The final results are displayed in Fig.7.

Fig. 6. Indicating the best geographical location for solar panels


Fig. 7. Simulation: a) simulation of MSX-64 solar panel; b) I-V curves for maximum performance of the MSX-64 panel at the optimal location in Kermanshah Province

According to the I-V curve in Fig.7b, the

maximum power point (MPP), established as the point of maximum voltage and current, was obtained at a VMP of 17.1 volts, IMP of 3.396 amperes and PMAX of 58 watts. 4.Conclusions The final result in terms of maximum power differs slightly from the 64 watt MSX-64 panel. Please note that the factor of greatest significance in regards to the decrease in

power is the low level of radiation in the province of Kermanshah during winter solstice. We may get the better results if we have more parameters, for example, number of clear days, cloudy hours, or else. If any more parameters are added to the regression equation, it is subject to change and it will be better. We only have the permission to access two kinds of data during periods 1995 to 2005. Therefore, we were forced to choose specific temperature and sunshine hours for this research.


Acknowledgements We would like to thank Islamic Azad University Kermanshah Branch for supporting this research. The author would like to thank anonymous reviewers for their constructive comments on this paper. References [1] Lund H., Mathiesen B. V., Energy System

Analysis of 100% Renewable Energy Systems,The Case of Denmark in Years 2030 and 2050, Energy 4th Dubrovnik Conference (2009) Jan.12-13, 34: 524–531.

[2] Nature of solar energy, http://www.suna.org.ir/fa/sun/nature, Accessed (2015)20 Jan.

[3] Woolfson M., The Origin and Evolution of the Solar System (2000) 1.12–1.19, ISBN 0750304588, Wiley.

[4] Nema S., Nema R. K., Agnihotri G., MATLAB/Simulink Based Study of Photovoltaic Cells/Modules/Array and Their Experimental Verification, International Journal of Energy and Environ (2010) 1: 487–500.

[5] Mulvaney D., Green Technology, an A-to-Z Guide (2011) 1–524, ISBN 1412996929, SAGE Publications Press.

[6] Femia N., Petrone G., Spagnuolo G., et al., Optimization of Perturb and Observe Maximum Power Point Tracking Method, IEEE Transactions on Power Electronics (2005) 20: 963–973.

[7] Faranda R., Leva S., Energy Comparison of MPPT Techniques for PV Systems, WSEAS Transactions on Power Systems (2008) 3: 446–455.

[8] Kumari J. S., Babu C. S., Mathematical Modeling and Simulation of Photovoltaic Cell Using Matlab-Simulink Environment, International Journal of Electrical and Computer Enginering (IJECE) (2011) 2: 26–34.

[9] Climatology & Geography of Kermanshah Province - Meteorological Organization Kermanshah’, http://www.kermanshahmet.ir/page.aspx?lang=fa-ir&id=b5ed122d-f465-4813-b596-54a0a1d2d895, Accessed (2014) 23 Jul.

[10]Meteorological Kermanshah’, http://www.kermanshahmet.ir/page.aspx?lang=fa-ir&id=b5ed122d-f465-4813-b596-54a0a1d2d895, Accessed (2015) 22 Jan.

[11] Statistics 200 synoptic stations in Iran’, http://www.chaharmahalmet.ir/iranarchive.asp, Accessed (2015)22 Jan.

[12] González-Longatt F. M., Model of Photovoltaic Module in Matlab, II CIBELEC Conference (2005) Jun, 2:1-5.

[13] Mishra B., Kar B. P., Matlab Based Modeling of Photovoltaic Array Characteristics, Bachelor thesis, National Institute of Technology, Rourkela, (2012).

[14] Pagliaro M., Ciriminna R., and Palmisano G., Flexible solar cells (2008) 880–891, ISBN 3527323759, ChemSusChem Press 1(11) Wiley.

[15] Sinton R. A., Forsyth M. K., Blum A. L., et al., Characterization of Substrate Doping and Series Resistance During Solar Cell Efficiency MEASUREMENT, United States Patent Application (US20140333319 A1) (2014) 1: 1–5.

[16] Bernardi M., Novel Materials, Computational Spectroscopy, and Multiscale Simulation in Nanoscale Photovoltaics, PhD thesis, Massachusetts Institute of Technology, Massachusetts (2013).

[17] Solarex MSX-64 Solar Panel, http://www.solarelectricsupply.com/solarex-msx-64-w-junction-box-548, Accessed (2015) 20 Jan.

[18] Rouholamini A., Pourgharibshahi H., Fadaeinedjad R., et al., Temperature of a Photovoltaic Module under the Influence of Different Environmental Conditions–Experimental Investigation, International Journal of Ambient Energy (2016) 37: 1–7.

[19] Visoly-Fisher I., Mescheloff A., Gabay M., et al., Concentrated Sunlight for Accelerated Stability Testing of Organic Photovoltaic Materials, Towards Decoupling Light Intensity and Temperature, Solar Energy Materials and Solar Cells (2015) 134: 99–107.

[20] Said S., Massoud A., Benammar M., et al., A Matlab/Simulink-Based Photovoltaic Array Model Employing SimPowerSystems Toolbox, Journal of Energy and Power Engineering (2012) 6: 1965–1975.

matlab simulation of solar panel msx-64 at the best...

Documents